Etiology, treatment outcome and prognostic factors among patients with secondary peritonitis at Bugando Medical Centre, Mwanza, Tanzania

Introduction Secondary peritonitis due to perforation of the gastrointestinal tract is one of the most common surgical emergencies all over the world and is associated with significantly morbidity and mortality. Previous studies conducted at Bugando Medical Centre (BMC) were retrospective and each was focused on single etiology; therefore there was an obvious need to evaluate the etiologies, treatment outcome and their prognostic factors altogether. Methods This was a descriptive cross-sectional study involving patients with secondary peritonitis admitted at BMC from May 2014 to April 2015. Sociodemographic and clinical characteristics among consented patients were collected using questionnaires. Peritoneal aspirate, biopsy and blood were collected perioperatively and processed using standard operating procedures. Analysis was done using STATA version 11 software. Results The study enrolled 97 patients with the female to male ratio of 1:1.8 and approximately 41.2 % (40/97) were in their third and fourth decades of life. Only 3 (3.09 %) patients arrived to the hospital within 24 hours of onset of illness, 26 (26.80 %) patients presented with shock and HIV seropositivity among all patients was 13.40 % (13/97). The common etiologies of secondary peritonitis were perforated appendicitis 23 (23.71 %), peptic ulcer disease 18 (18.56 %), ischemia 18 (18.56 %) and typhoidal perforation 15 (15.46 %). Of the 97 patients, 35 (36.08 %) had complications and 15 (15.46 %) died. Presence of premorbid illness and post-operative complication were found to be associated with death (p values = 0.004 and <0.001 respectively). Conclusions The most common etiologies of secondary peritonitis at BMC are perforated appendicitis, peptic ulcer disease, ischemia and typhoidal perforation. Premorbid illness and postoperative complications in this setting are associated with death and as the matter of fact proper screening on admission should be done to identify patients with premorbid illness and confer prompt management.


Introduction
Peritonitis has a long historical background and is conventionally divided into three broad groups namely; primary, secondary and tertiary basing on the source and nature of the microbial contamination [1]. Secondary peritonitis which is due to perforation of the gastrointestinal tract is one of the most common surgical emergencies all over the world and it is associated with significantly morbidity and mortality [1][2][3][4]. Due to the loss of epithelial integrity, bacterial pathogens can traverse into the peritoneal cavity leading to a cascade of inflammatory response, sepsis, multisystem organ failure and death if not treated in a timely manner [1]. Aggressive resuscitation and early surgical intervention is therefore required in order to deliver optimal care for the patients and improve their treatment outcome [2][3][4][5].
The spectrum of gastrointestinal perforation have a wide geographical variations; with preponderance of lower gastrointestinal perforations western countries as opposed to upper gastrointestinal perforations in developing countries [3,[6][7][8].
At Bugando Medical Centre (BMC), primary bacterial peritonitis was recently reported to have a prevalence of 11.6 % among patients with portal hypertension and ascites [9]. Moreover, secondary peritonitis is the commonest indication for admission in the surgical wards resulting into increased workload, increased duration of hospital stays, and complications such as enterocutaneous fistula, surgical site infections and sepsis. Furthermore, the death rates attributable to secondary peritonitis in this settings has been shown to be 10.7 % and 23.1 % among patients with peptic ulcer and typhoid perforations respectively [10][11][12]. Despite advances in surgical techniques, antimicrobial therapy, and intensive care support; the management of peritonitis continues to be highly demanding, difficult, and complex [3,13]. Early prognostic evaluation of peritonitis is desirable to provide objective classification of the severity of the disease and select highrisk patients for more aggressive therapeutic procedures [14,15]; but this has been affected by late presentation to the health facilities by majority of patients, a situation which further complicates effective management [16].
Previous studies from BMC were retrospective and each focused on single etiology thus, lacking a holistic approach in addressing the implicated etiologies [10][11][12]; therefore the present study aimed at exploring the wide range of etiologies causing secondary peritonitis so as to identify factors responsible for the poor outcome and specifically address them in the context of reducing poor outcomes among patients with secondary peritonitis.

Study design, site and sampling procedures
A descriptive cross-sectional study involving patients with secondary peritonitis admitted at BMC from May 2014 to April 2015 inclusive was conducted. BMC is one of the four largest referral hospitals in the country, located in the northwestern part of Tanzania. It has a bed capacity of 1000 and serves a catchment population of approximately 13 million people. It is also a consultant and teaching hospital for the Catholic University of Health and Allied Sciences-Bugando (CUHAS). A minimum sample size of 93 was estimated basing on the Yamane Taro (1967) using previous local data which revealed 243 patients who were operated with secondary peritonitis in 2013 as well as accepted error of 8 %. Patients who met the inclusion criteria were offered explanations about the study and requested to provide a written informed consent before being enrolled into the study.
Preoperatively, all the patients recruited into the study were resuscitated with intravenous fluids to correct fluid and electrolyte deficits; nasogastric suction; urethral catheterization and broad-spectrum antibiotic coverage as per BMC protocol. Relevant preoperative investigations included hemoglobin estimation, blood grouping and cross-matching. In addition, provider initiated testing for HIV infection was offered to all patients according to National AIDS Control Program [17,18]. For patients found to be HIV seropositive, CD4+ count using FACSCALIBUR (BD Biosciences, USA) was done. Radiological investigations included abdominal x-rays (erect and supine) as well as abdominal ultrasound were performed in all patients.
After resuscitation, all patients had to undergo preoperative anesthetic assessment using the American Society of Anesthetists (ASA) classification [19]. Patients who were clinically and hemodynamically stable for surgery were subjected to exploratory laparotomy under general anesthesia by the residents using the standard techniques.
The diagnosis of secondary peritonitis was established clinically based on guarding, rigidity, and tenderness on palpation of the abdomen, supported radiologically by detection of free air under the diaphragm and/or free intra peritoneal fluid on abdominal ultrasound and confirmed by intraoperative clinical and investigational findings.
Intraoperatively, purulent materials were aspirated from peritoneal cavity for Ziehl-Neelsen stain for Acid fast bacilli (AFB), tissue biopsy and regional lymph nodes were taken for histopathological analysis [20]. The primary outcome of the study was survival i.e. whether a patient died or was alive at the time of discharge.

Data management
A structured questionnaires were used to collect sociodemographic and clinical data preoperatively, intraoperatively and post-operatively. Data were entered into Microsoft excel and then exported to the STATA version 11.0 software for analysis according to the objectives of the study. The results are presented into percentages/ proportions for categorical variables whereas continuous variables are described as mean (± standard deviation) or median (interquantile range) depending on the distribution of data. The difference in distribution of a predictor variable was considered significant if p-value is less than 0.05.

Ethical considerations
Ethical clearance was obtained from the joint BMC/ CUHAS ethical committee. Written informed consent was obtained from every participant prior to be involved in the study. Anonymity of patients was ensured by use of codes. Laboratory results were timely reported to the attending doctors for specific management.

Results
A total of 112 patients were diagnosed to have secondary peritonitis and were admitted during the study period; of these 97 fulfilled the inclusion criteria and were enrolled whereas15 patients were excluded (samples were not taken in 10 and 5 did not fulfill inclusion criteria).
The female to male ratio of the study participants was 1:1.8. The median age (IQR) of the study was 32 (21-47) years, the youngest was 5 years old and the oldest was 86 years old. Moreover, approximately 41.2 % (40/97) were in their third and fourth decades of life, 51 (52.58 %) were from urban, and majority had primary school education (59.79 %) and were unemployed (62.89 %).

Clinical presentations
Majority of patients (96.91 %, 94/97) presented to hospital more than 24 hours since the illness started. Twenty seven (27.84 %) patients presented with co morbid illness, with HIV infection being the predominant co morbidity in about 46.43 % (13/28) of patients. Abdominal pain was found to be the most common presenting symptom in about 95 (97.6 %) patients and 93 (95.88 %) were diagnosed with generalized peritonitis. About 26 (26.80 %) patients presented with shock with systolic blood pressure bellow 90 mmHg and heart rate was above 100 beats per minute and 56 (57 .73 %) patients were febrile (Table 1).

Laboratory and radiological investigations
The median hemoglobin (IQR) of the 97 patients was 9.2 (7.3-11) gm/dl; ranging from 1.2 gm/dl to 17 gm/dl. Widal test was done in 71 patients and titers were high in 14 patients and were within normal range in 57 patients. Thirteen patients (13.40 %) were found to be HIV seropositive; with the median CD4 count of 250 (150-330) and lowest level was 95 and highest was 480.

The prognostic factors for mortality
There was no significant association between sociodemographic characteristics of patients with the outcome.
Presence of premorbid illness and post-operative complications were found to be significantly associated with mortality (p value of 0.004 and <0.001 respectively) ( Tables 5 and 6).

Discussion
The etiology of secondary peritonitis at BMC The predominance of perforated appendicitis, peptic ulcer disease, ischemia and typhoidal perforation as the common etiologies of secondary peritonitis in this study are similar to the study done at St. Francis Hospital Nsambya Kampala-Uganda [21] and other two previous studies in the same hospital [10,11]. Ruptured appendix and ischemia leading to gangrenous sigmoid volvulus may be due to poor health seeking behavior, as shown in this study whereby majority of patients (96 %) came to hospital more than 24 hours after the onset of illness or mismanagement in the lower health facilities. The reason for ruptured peptic ulcer and typhoidal perforations may be due to poor sanitary condition in developing countries, exposing the patients to Salmonella infection and Helicobacter pylori infection in the community, which in turn may result into perforations.

The treatment outcome of secondary peritonitis
Similar to another previous study [22], about 36.08 % of patients with secondary peritonitis in this study developed complications namely superficial surgical site infection, wound dehiscence and septicemia. Mortality rate was found to be 15.46 % and median length of hospital stay was 7 days. These findings are comparable to other studies [8,[22][23][24] but lower than what were reported in the same hospital four years ago by Chalya et al. [11] which showed complication rate to be 39.4 %, superficial surgical site infection 55.5 %, mortality 23.1 % and median length of hospital stay of 28 days in patients with typhoidal perforation [11]. The difference could be related to the recently renovated modern accident & emergency department and ICU which are well equipped     (18.55 %) is similar to what is expected in contaminated wounds as reported previously (21 %); but lower than 26 % reported before in the same hospital by Mawala et al. [1,25]. This could be due to the fact when the previous study was conducted, the department of surgery consisted of general surgery, urology, orthopedic & traumatology as well as otorhinolaryngology which are now separate departments, and thus more care can be offered timely as a result of this decentralization.

The prognostic factors for mortality among patients with secondary peritonitis
In this study, only premorbid illness and post-operative complications were statistically found to predict deaths. This is possible because less priority can be given to other premorbid illness in case of obvious surgical condition of the patient and thus, these premorbid illnesses may be overlooked resulting into poor outcome [19,[25][26][27]. As a result, the present study emphasizes on carefully screening of potential premorbid illnesses and also alert care providers to ensure that they make all efforts to prevent postoperative complications. The predominance of HIV among premorbid illnesses in this and previous study in the same hospital [28] worthy to be further explored using a long term prospective study with large sample size so as to specifically delineate the contribution of HIV in the outcome of patients with secondary peritonitis. But from this study, it is sufficing to reiterate that approximately 30.8 % of patients with HIV infections died and their median CD4+ count were relatively low (172.5 cells/ mm 3 ) compared to HIV infected patients who survived (315 cells/mm 3 ). Development of complications postoperatively was strongly associated with death in the present study, similarly another study revealed that development of complication like sepsis can escalate mortality [22]. Sociodemographic factors such as age, gender, clinical symptoms such as vomiting and sign like fever in this study and another related study were not found to predict deaths [23].

The limitation of the study
The present study diagnosed peptic ulcer disease basing of clinical (site of perforation) and histology. Stool antigen test for Helicobacter pylori could have been of interest to ascertain the role of this point-of-care test and this is recommended in future studies.

Conclusions
The most common etiologies of secondary peritonitis at BMC are perforated appendicitis, peptic ulcer disease, ischemia and typhoid. Premorbid illnesses and postoperative complications in this setting are associated with death and as the matter of fact proper screening on admission should be done to identify patients with premorbid illness and confer prompt management to prevent postoperative complications and subsequent death. Moreover, a study to explore the long-term outcomes of patients with secondary peritonitis is recommended.
Submit your next manuscript to BioMed Central and take full advantage of: